An improved segmentation method for porous transducer CT images

Meiling Wang, Ruoyu Guo, Ke Ning, Li Ming

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The paper presents an improved image segmentation method with a straightforward workflow for porous transducer CT images, which can be used to establish porous transducer three-dimensional model and further study its characteristics. Data distribution of CT images is firstly analyzed and Gaussian filtering is conducted to reduce divergence of CT images. An improved fully convolutional neural network model based on U-Net, for which multi-channel images are set as network input, is trained using training set. The proposed method improves pore connectivity of the segmentation results. Improvement of porosity and permeability relative errors as well as MIOU on test set shows that the proposed method is an effective and generic two-phase segmentation method for porous transducer CT images without need of adjusting any parameters.

Original languageEnglish
Title of host publicationEleventh International Conference on Digital Image Processing, ICDIP 2019
EditorsJenq-Neng Hwang, Xudong Jiang
PublisherSPIE
ISBN (Electronic)9781510630758
DOIs
Publication statusPublished - 2019
Event11th International Conference on Digital Image Processing, ICDIP 2019 - Guangzhou, China
Duration: 10 May 201913 May 2019

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume11179
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

Conference11th International Conference on Digital Image Processing, ICDIP 2019
Country/TerritoryChina
CityGuangzhou
Period10/05/1913/05/19

Keywords

  • Fully convolutional neural network
  • Image segmentation
  • Permeability
  • Pore connectivity
  • Porous transducer

Fingerprint

Dive into the research topics of 'An improved segmentation method for porous transducer CT images'. Together they form a unique fingerprint.

Cite this